Objective: To assess the effectiveness of antenatal screening for Down's syndrome by maternal age and routine mid-pregnancy ultrasound scanning.
Design: Retrospective six year survey.
Setting: Maternity units of a district general hospital.
Subjects: Pregnant women booked for delivery in hospital between 1 January 1993 and 31 December 1998.
Main outcome measures: All cases of Down's syndrome occurring in district identified from regional congenital anomaly register and cytogenetic laboratory records. Women's case notes were examined to identify indication for karyotyping, gestation at diagnosis, and outcome of pregnancy.
Results: 31 259 deliveries occurred during study period, and 57 cases of Down's syndrome were identified, four in failed pregnancies and 53 in ongoing pregnancies or in neonates. The analysis was confined to ongoing pregnancies or liveborn children. Invasive antenatal tests were performed in 6.6% (2053/31 259), and 68% (95% confidence interval 56% to 80%) of cases of Down's syndrome were detected antenatally, giving a positive predictive value of 1.8%. There were 17 undetected cases, and in seven of these the women had declined an offer of invasive testing. In women aged less than 35 years the detection rate was 53% (30% to 76%). Most of the cases detected in younger women followed identification of ultrasound anomalies.
Conclusions: The overall detection rate was considerably higher than assumed in demonstration projects for serum screening. As a result, the benefits of serum screening are much less than supposed. Before any new methods to identify Down's syndrome are introduced, such as nuchal translucency or first trimester serum screening, the techniques should be tested in properly controlled trials.